Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A non-transitory computer-readable medium storing computer-executable instructions, that when executed by at least one processor of a computing device, cause the computing device to determine a promotion price schedule to be charged for each item in a group of items by a retailer computing device that is remote from the computing device by: receiving, with the at least one processor, a set of promotion prices for each item in the group, wherein the set of promotion prices limits permissible promotion prices to be charged by the retailer computing device for the item to a discrete number of prices that are input to the computing device; with the at least one processor, and for each item, each time period in the price schedule, and each promotion price in the set for the item, determining an item coefficient that corresponds to a change in a value of an objective function when the item is priced at the promotion price, where the objective function is based on a non-linear demand model that is a function of i) past prices of the item charged by the retailer computing device during a previous time period that preceded a current time period, and ii) current prices of different items in the group being charged by the retailer computing device during the current time period; formulating, with the at least one processor, an approximate objective function that: (i) is a linear approximation of the non-linear demand model based on the discrete number of prices included in the set of promotion prices, and (ii) includes a sum of products, where each product includes at least one of the item coefficients multiplied by a binary item decision variable; transmitting, with the at least one processor, the item coefficients, the approximate objective function, and one or more constraints to an optimizer that is configured to determine values of the item decision variables that maximize the approximate objective function, wherein the one or more constraints comprise a cross no-touch constraint that defines at least a minimum period of time that is to elapse between time periods when the promotion price for different items in the group can be offered; generating, with the at least one processor, the promotion price schedule for each item based on at least values of the item decision variables determined by the optimizer, and generating a data structure stored in memory containing the promotion price schedule; and controlling, with the at least one processor, the retailer computing device by transmitting the data structure containing the promotion price schedule to the retailer computing device over a network connection, causing the retailer computing device to modify prices charged by the retailer computing device for a plurality of the items in the group to the promotion prices in the promotion price schedule.
This invention relates to dynamic pricing optimization for retail items using a non-linear demand model. The system determines promotion price schedules for a group of items by first receiving a set of permissible promotion prices for each item, which limits the possible prices to a discrete set. For each item, time period, and promotion price, the system calculates an item coefficient representing the impact of pricing the item at that promotion price on an objective function. The objective function is based on a non-linear demand model that considers past prices of the item and current prices of related items. The system then formulates an approximate objective function as a linear approximation of the non-linear demand model, incorporating the item coefficients and binary decision variables. This approximation is transmitted to an optimizer, which determines the optimal values of the decision variables to maximize the objective function while adhering to constraints, including a cross no-touch constraint that enforces a minimum time interval between promotions for different items. The resulting promotion price schedule is generated and transmitted to a retailer's computing system, which adjusts the prices of the items accordingly. The system enables retailers to optimize pricing strategies while accounting for complex demand interactions and ensuring compliance with promotional constraints.
2. The non-transitory computer-readable medium of claim 1 , further comprising instructions, that when executed by the at least one processor, cause the computing device to determine the promotion price schedule for each item by: computing a pair coefficient for each pair of items, each time period in the price schedule, and each combination of two of promotion prices, where a first promotion price in the combination is selected from the set of promotion prices for a first item in the pair and a second promotion price in the combination is selected from the set of promotion prices for a second item in the pair, further where the pair coefficient corresponds to a change in a value of the objective function when the first item is priced at the first promotion price and the second item is priced at the second promotion price; formulating the approximate objective function such that the approximate objective function includes a sum of products, where each product includes a pair coefficient multiplied by a binary pair decision variable; providing the pair coefficients with the item coefficients, the approximate objective function, and one or more constraints to the optimizer, where the optimizer is configured to determine, as constrained by the one or more constraints, values of the item decision variables and the pair decision variables that maximize the approximate objective function; and creating the promotion price schedule for each item based on at least the values of the item decision variables and the pair decision variables determined by the optimizer.
This invention relates to dynamic pricing optimization for retail promotions, specifically improving computational efficiency in determining optimal promotion price schedules for multiple items. The problem addressed is the high computational complexity of solving large-scale pricing optimization problems where interactions between items (e.g., cannibalization or complementarity effects) must be considered. The solution involves approximating a complex objective function that accounts for both individual item pricing decisions and pairwise item interactions. For each pair of items, the system computes a "pair coefficient" representing the impact on the objective function (e.g., revenue or profit) when specific promotion prices are applied to both items. These coefficients are used to construct an approximate objective function as a sum of products, where each product combines a pair coefficient with a binary decision variable indicating whether the corresponding price pair is selected. The system then provides these coefficients, along with item-specific coefficients and constraints (e.g., budget limits or minimum/maximum prices), to an optimizer. The optimizer determines values for both item-level and pair-level decision variables that maximize the approximate objective function while respecting constraints. The final promotion price schedule is generated based on these optimized values. This approach reduces computational complexity by leveraging pairwise interactions without requiring exhaustive enumeration of all possible price combinations.
3. The non-transitory computer-readable medium of claim 2 , further comprising instructions, that when executed by the at least one processor, cause the computing device to create the promotion price schedule for an item by: when an item decision variable for a period has a value of zero, assigning the item's price in the period to a non-promotion price for the item; when an item decision variable for a period has a value of one, assigning the item's price in the period to a promotion price associated with the item decision variable; and when a pair decision variable for a period has a value of one, assigning the first item's price in the period to the first promotion price associated with the pair decision variable and assigning the second item's price in the period to the second promotion price associated with the pair decision variable.
This invention relates to dynamic pricing systems for retail or e-commerce platforms, specifically addressing the challenge of optimizing pricing strategies to maximize revenue or profit. The system uses decision variables to determine whether an item should be priced at a non-promotion price, a standalone promotion price, or as part of a paired promotion with another item. The decision variables are binary, where a value of zero indicates no promotion, and a value of one triggers a promotion. For standalone promotions, the item's price is set to a predefined promotion price. For paired promotions, the system assigns distinct promotion prices to two items, allowing for bundled or complementary pricing strategies. The system dynamically generates a promotion price schedule based on these decision variables, enabling flexible and data-driven pricing adjustments. This approach helps retailers optimize pricing to influence customer behavior, clear inventory, or maximize margins while maintaining control over individual or paired item pricing. The invention is implemented via a computing device executing instructions stored on a non-transitory computer-readable medium, ensuring scalability and real-time adaptability.
4. The non-transitory computer-readable medium of claim 1 , where the one or more constraints further comprise a maximum number of items in the group that are assigned a promotion price during a same period.
This invention relates to dynamic pricing systems for retail or e-commerce platforms, specifically addressing the challenge of optimizing promotional pricing while managing inventory and revenue constraints. The system automatically assigns promotion prices to items within a predefined group, such as a product category or inventory segment, to maximize sales or revenue. A key constraint is limiting the number of items in the group that can receive a promotion price simultaneously, ensuring that promotions do not deplete inventory too quickly or negatively impact profit margins. The system evaluates real-time data, such as demand trends, inventory levels, and competitor pricing, to determine which items should receive promotions and for how long. By enforcing this constraint, the system prevents over-promotion of certain items while still driving sales growth. The approach balances promotional effectiveness with operational and financial stability, making it suitable for large-scale retail environments where dynamic pricing is critical. The invention improves upon traditional static pricing models by incorporating adaptive constraints that respond to market conditions.
5. The non-transitory computer-readable medium of claim 1 , where the one or more constraints further specifies a must-promote set of items, wherein if one item in the must-promote set is assigned a promotion price in a given period, all other items in the must-promote set must also be assigned a promotion price during a same period.
This invention relates to a system for managing promotional pricing of items in a retail or e-commerce environment. The problem addressed is ensuring consistent promotion of related items, such as bundled products or complementary goods, to maintain pricing fairness and customer expectations. The solution involves a constraint-based approach where a "must-promote" set of items is defined. If any item in this set is assigned a promotional price during a specific period, all other items in the set must also receive a promotional price during the same period. This ensures that related items are promoted together, preventing scenarios where only some items in a group are discounted, which could lead to customer confusion or dissatisfaction. The system enforces this rule through automated pricing logic, ensuring compliance with the defined constraints. The invention is implemented via a non-transitory computer-readable medium containing instructions for executing the pricing logic, which may include additional constraints such as minimum or maximum promotion durations, eligible discount ranges, or exclusion rules for certain items. The system dynamically adjusts promotions to meet these requirements while optimizing pricing strategies. This approach improves promotional consistency, enhances customer trust, and simplifies pricing management for retailers.
6. The non-transitory computer-readable medium of claim 1 , where the one or more constraints further specifies a cannot-promote set of items, wherein if one item in the cannot-promote set is assigned a promotion price in a given period, none of the other items in the cannot-promote set may be assigned a promotion price during a same period.
This invention relates to dynamic pricing systems for retail or e-commerce platforms, specifically addressing conflicts in promotional pricing strategies. The problem solved is the unintended competition or cannibalization of sales when multiple related or similar items are simultaneously promoted, leading to reduced overall revenue or customer confusion. The system enforces constraints to prevent such conflicts by defining a "cannot-promote" set of items. If one item in this set is assigned a promotional price during a given time period, no other item in the set can be promoted during the same period. This ensures that promotions are strategically managed to avoid undermining each other. The system may also include additional constraints, such as budget limits, promotional frequency rules, or competitive pricing thresholds, to further optimize pricing decisions. The invention is implemented via a non-transitory computer-readable medium containing instructions for executing these constraints within a pricing algorithm. The solution is particularly useful for retailers managing large product catalogs where overlapping promotions could dilute marketing impact or reduce profitability.
7. The non-transitory computer-readable medium of claim 1 , where the one or more constraints further comprise a fixed relationship between prices of given items in the group that must be maintained in all periods.
This invention relates to a system for managing item pricing in a dynamic pricing environment, particularly where maintaining specific price relationships between items is critical. The problem addressed is ensuring that certain items within a group must retain a fixed price ratio or relationship across all pricing periods, even as other pricing factors (such as demand, competition, or inventory levels) fluctuate. This is important in industries where relative pricing must be preserved, such as bundled products, subscription tiers, or complementary goods. The system involves a non-transitory computer-readable medium storing instructions that, when executed, enforce pricing constraints. These constraints include a fixed relationship between the prices of given items in a group, meaning that the ratio or proportional difference between their prices must remain constant over time. For example, if Item A must always be priced at twice the cost of Item B, the system ensures this relationship holds regardless of external pricing adjustments. The constraints may also include other pricing rules, such as minimum or maximum price thresholds, or time-based pricing adjustments. The system dynamically adjusts prices while ensuring the fixed relationship is maintained, preventing pricing conflicts or deviations that could disrupt business strategies or customer expectations. This approach is particularly useful in automated pricing systems where manual oversight is impractical, ensuring consistency and compliance with predefined pricing policies. The solution is applicable in e-commerce, subscription services, and other markets where relative pricing stability is essential.
8. A computing system configured to determine a promotion price schedule to be charged for each item in a group of items by a retailer computing device that is remote from the computing device, the computing system comprising: at least one processor connected to at least one memory; a non-transitory computer readable medium operably connected to the at least one processor and configured with a formulation logic configured with instructions that when executed by the at least one processor, cause the at least one processor to: receive, a set of promotion prices for each item in the group, wherein the set of promotion prices limits permissible promotion prices to be charged by the retailer computing device for the item to a discrete number of prices that are input to the computing system; for each item, each time period in the price schedule, and each promotion price in the set for the item, generate an item coefficient that corresponds to a change in a value of an objective function when the item is priced at the promotion price, where the objective function is based on a non-linear demand model that is a function of i) past prices of the item charged by the retailer computing device during a previous time period that preceded a current time period, and ii) current prices of different items in the group being charged by the retailer computing device during the current time period; formulate an approximate objective function that: (i) is a linear approximation of the non-linear demand model based on the discrete number of prices included in the set of promotion prices, and (ii) includes a sum of products, where each product includes at least one of the item coefficients multiplied by a binary item decision variable; transmit the item coefficients, the approximate objective function, and one or more constraints to an optimizer that is configured to determine values of the item decision variables that maximize the approximate objective function, wherein the one or more constraints comprise a cross no-touch constraint that defines at least a minimum period of time that is to elapse between time periods when the promotion price for different items in the group can be offered; and solution logic stored on the non-transitory computer readable medium and configured with instructions that when executed by the at least one processor, cause the at least one processor to: receive, from the optimizer, the values of the item decision variables; generate a promotion price schedule for each item based on at least the values of the item decision variables; and control the retailer computing device by transmitting the promotion price schedule to the retailer computing device over a network connection, to cause the retailer computing device to modify prices charged by the retailer computing device for a plurality of the items in the group to the promotion prices in the promotion price schedule.
A computing system determines optimal promotion price schedules for items in a group, ensuring prices align with retailer constraints and maximize revenue. The system receives a predefined set of permissible promotion prices for each item, limiting prices to a discrete number of options. For each item, time period, and promotion price, the system generates item coefficients representing the impact of pricing decisions on an objective function. This function is based on a non-linear demand model that considers past prices of the item and current prices of related items. The system then formulates a linear approximation of this model, using item coefficients and binary decision variables to simplify optimization. Constraints, including a cross no-touch rule, enforce minimum time gaps between promotions for different items. The system transmits these coefficients, the approximate objective function, and constraints to an optimizer, which determines the best promotion prices. The system then generates a price schedule and transmits it to a remote retailer computing device, updating the prices accordingly. This approach ensures dynamic, constraint-compliant pricing that maximizes revenue while accounting for demand sensitivity and competitive pricing.
9. The computing system of claim 8 , where the solution logic is further configured to: when an item decision variable for a period has a value of zero, assign the item's price in the period to a non-promotion price for the item; and when an item decision variable for a period has a value of one, assign the item's price in the period to a promotion price associated with the item decision variable.
This invention relates to computing systems for dynamic pricing and promotion management in retail or e-commerce environments. The system addresses the challenge of optimizing pricing strategies by automatically adjusting item prices based on predefined decision variables, which can indicate whether an item is promoted or sold at a regular price during a given period. The computing system includes solution logic that processes item decision variables for each time period. If the decision variable for an item in a specific period is zero, the system assigns the item's price to a non-promotion (regular) price. Conversely, if the decision variable is one, the system assigns the item's price to a promotion price associated with that decision variable. This logic ensures that pricing adjustments are systematically applied based on predefined conditions, enabling automated promotion management and dynamic pricing strategies. The system may also include a data store containing item pricing data, including regular and promotional prices, and a processor to execute the solution logic. The decision variables can be determined by external algorithms, user inputs, or other business rules, allowing flexibility in defining promotion criteria. The invention improves efficiency in pricing management by automating the application of promotional discounts while maintaining control over pricing strategies.
10. The computing system of claim 8 , where: the formulation logic is further configured to: compute a pair coefficient for each pair of items, each time period in the price schedule, and each combination of two of promotion prices, where a first promotion price in the combination is selected from the set of promotion prices for a first item in the pair and a second promotion price in the combination is selected from the set of promotion prices for a second item in the pair, further where the pair coefficient corresponds to a change in a value of the objective function when the first item is priced at the first promotion price and the second item is priced at the second promotion price; formulate the approximate objective function such that the approximate objective function includes a sum of products, where each product includes a pair coefficient multiplied by a binary pair decision variable; provide the pair coefficients with the item coefficients, the approximate objective function, and one or more constraints to the optimizer, where the optimizer is configured to determine values of the item decision variables and the pair decision variables that maximize the approximate objective function; and the solution logic is further configured to: receive, from the optimizer, the values of the item decision variables and the pair decision variables; and create the promotion price schedule each item based on at least the values of the item decision variables and the pair decision variables.
This invention relates to a computing system for optimizing promotion pricing in retail or e-commerce environments. The system addresses the challenge of determining optimal promotion prices for multiple items to maximize business objectives, such as revenue or profit, while accounting for interactions between items. Traditional pricing optimization often fails to consider how promoting one item may influence demand for another, leading to suboptimal pricing strategies. The system includes formulation logic that computes pair coefficients for every possible combination of two items, each time period in the price schedule, and each possible pair of promotion prices. Each pair coefficient quantifies the impact on an objective function (e.g., revenue or profit) when one item is priced at a specific promotion price and another item is priced at another promotion price. The formulation logic then constructs an approximate objective function that includes a sum of products, where each product is a pair coefficient multiplied by a binary decision variable representing whether the price pair is selected. The system also computes item coefficients for individual items, representing their standalone impact on the objective function. An optimizer receives the pair coefficients, item coefficients, approximate objective function, and constraints (e.g., budget limits or minimum/maximum prices). The optimizer determines the values of item and pair decision variables that maximize the objective function. Solution logic then generates a promotion price schedule for each item based on these optimized values, ensuring that the final pricing strategy accounts for both individual item performance and cross-item interactions. This approach improves pricing decisions by leveraging interdepe
11. The computing system of claim 10 , where the solution logic is further configured to, when a pair decision variable for a period has a value of one, assign the first item's price in the period to the promotion price associated with the pair decision variable and assigning the second item's price in the period to the promotion price associated with the pair decision variable.
A computing system is designed to optimize pricing strategies for items in a retail or e-commerce environment, particularly for promotions involving pairs of items. The system addresses the challenge of dynamically adjusting prices to maximize revenue or other business objectives when items are promoted together. The system includes solution logic that evaluates decision variables associated with item pairs to determine optimal pricing. When a decision variable for a specific time period indicates a promotion (value of one), the system assigns the price of the first item in the pair to a promotion price and similarly assigns the price of the second item to the same promotion price. This ensures consistent pricing for both items during the promotion period. The system may also include a data store for storing item and pricing data, as well as a processor for executing the solution logic. The solution logic may further be configured to handle multiple decision variables and time periods, allowing for flexible and adaptive pricing strategies. The system aims to improve promotional effectiveness by dynamically adjusting prices based on predefined decision variables, enhancing revenue or other business metrics.
12. A computer-implemented method for determining a promotion price schedule to be charged for each item in a group of items by a retailer computing device that is remote from a computing device, the method comprising: receiving, a set of promotion prices for each item in the group with the computing device, wherein the set of promotion prices limits permissible promotion prices to be charged by the retailer computing device for the item to a discrete number of prices that are input to the computing device; for each item, each time period in the price schedule, and each promotion price in the set for the item, determine an item coefficient that corresponds to a change in a value of an objective function when the item is priced at the promotion price, where the objective function is based on a non-linear demand model that is a function of i) past prices of the item charged by the retailer computing device during a previous time period that preceded a current time period, and ii) current prices of different items in the group being charged by the retailer computing device during the current time period; formulating an approximate objective function that: (i) is a linear approximation of the non-linear demand model based on the discrete number of prices included in the set of promotion prices, and (ii) includes a sum of products, where each product includes at least one of the item coefficients multiplied by a binary item decision variable; transmitting the item coefficients, the approximate objective function, and one or more constraints to an optimizer that is configured to determine values of the item decision variables that maximize the approximate objective function, wherein the one or more constraints comprise a cross no-touch constraint that defines at least a minimum period of time that is to elapse between time periods when the promotion price for different items in the group can be offered; generating, a promotion price schedule for each item based on at least values of the item decision variables determined by the optimizer, and generating a data structure stored in memory containing the promotion price schedule; and control the retailer computing device by transmitting the data structure containing the promotion price schedule to the retailer computing device over a network connection, causing the retailer computing device to modify prices of items in data records according to the corresponding values in the promotion price schedule.
This invention relates to dynamic pricing optimization for retailers, specifically determining promotion price schedules for groups of items while adhering to business constraints. The problem addressed is optimizing item pricing to maximize revenue or other objectives while respecting promotional pricing rules and avoiding customer fatigue from frequent price changes. The method involves a computing device receiving predefined promotion price options for each item in a group, where these options are limited to a discrete set of permissible prices. For each item, time period, and promotion price, the system calculates an item coefficient representing the impact of that price on an objective function (e.g., revenue). The objective function is based on a non-linear demand model that considers past prices of the item and current prices of related items. To simplify optimization, the system approximates this non-linear model as a linear function using the discrete price options, where the approximation includes products of item coefficients and binary decision variables. The system then transmits these coefficients, the approximate objective function, and constraints (including a "cross no-touch" rule enforcing minimum time gaps between promotions for different items) to an optimizer. The optimizer determines the optimal promotion prices by solving for the binary decision variables that maximize the approximate objective function. The resulting promotion price schedule is stored in a data structure and transmitted to a retailer's computing system, which updates item prices accordingly. This approach enables efficient, constraint-aware pricing optimization for large item groups.
13. The computer-implemented method of claim 12 , further comprising: computing a pair coefficient for each pair of items, each time period in the price schedule, and each combination of two of promotion prices, where a first promotion price in the combination is selected from the set of promotion prices for a first item in the pair and a second promotion price in the combination is selected from the set of promotion prices for a second item in the pair, further where the pair coefficient corresponds to a change in a value of the objective function when the first item is priced at the first promotion price and the second item is priced at the second promotion price; formulating the approximate objective function such that the approximate objective function includes a sum of products, where each product includes a pair coefficient multiplied by a binary pair decision variable; providing the pair coefficients with the item coefficients, the approximate objective function, and one or more constraints to the optimizer, where the optimizer is configured to determine, as constrained by the one or more constraints, values of the item decision variables and the pair decision variables that maximize the approximate objective function; and creating the promotion price schedule for each item based on at least the values of the item decision variables and the pair decision variables determined by the optimizer.
This invention relates to optimizing promotion pricing for items in a retail or e-commerce setting. The problem addressed is determining optimal promotion prices for multiple items over time to maximize a business objective, such as revenue or profit, while accounting for interactions between items. The solution involves computing pair coefficients for each pair of items, each time period, and each combination of promotion prices. These coefficients quantify the impact on an objective function (e.g., revenue or profit) when one item is priced at a specific promotion price and another item is priced at another promotion price. The method then formulates an approximate objective function that includes a sum of products, where each product is a pair coefficient multiplied by a binary pair decision variable. This approximate function is provided to an optimizer along with item coefficients, constraints, and the objective function. The optimizer determines values for item and pair decision variables that maximize the objective function while respecting constraints. Finally, a promotion price schedule is created for each item based on the optimized decision variables. This approach improves upon traditional pricing methods by explicitly modeling item interactions, leading to more accurate and profitable promotion strategies.
14. The computer-implemented method of claim 13 , further comprising: when an item decision variable for a period has a value of zero, assigning the item's price in the period to a non-promotion price for the item; when an item decision variable for a period has a value of one, assigning the item's price in the period to a promotion price associated with the item decision variable; and when a pair decision variable for a period has a value of one, assigning the first item's price in the period to the first promotion price associated with the pair decision variable and assigning the second item's price in the period to the second promotion price associated with the pair decision variable.
The invention relates to dynamic pricing systems for retail or e-commerce platforms, specifically addressing the challenge of optimizing pricing strategies to maximize revenue or profit. The system uses decision variables to determine whether items should be priced at regular or promotional rates, including scenarios where pairs of items are jointly promoted. When an item decision variable for a given period is zero, the item's price is set to its non-promotional value. If the item decision variable is one, the price is adjusted to a predefined promotion price. For paired items, when a pair decision variable is one, the system assigns distinct promotion prices to each item in the pair. The method ensures flexible pricing adjustments based on predefined decision variables, allowing for strategic promotions while maintaining control over individual or bundled item pricing. The approach optimizes revenue by dynamically applying promotions to single items or coordinated pairs, enhancing pricing strategies in competitive markets.
15. The computer-implemented method of claim 12 , where the one or more constraints further comprise a maximum number of items in the group that are assigned a promotion price during a same period.
This invention relates to dynamic pricing systems for assigning promotional prices to items within a group, particularly in retail or e-commerce environments. The problem addressed is the need to control the distribution of promotional prices to prevent over-discounting, which can reduce profitability or create unintended competitive disadvantages. The solution involves applying constraints to the pricing process to ensure that only a limited number of items in a group receive a promotion price during a specified time period. This constraint helps maintain pricing stability while still allowing selective discounts to drive sales. The method involves analyzing item data, such as historical sales, inventory levels, or customer demand, to determine which items should receive a promotion price. The system then applies the constraint to ensure that the number of promoted items does not exceed a predefined maximum during the same period. This approach balances promotional effectiveness with financial sustainability, preventing excessive discounts that could harm revenue. The invention is particularly useful in large-scale retail operations where automated pricing adjustments are necessary to manage thousands of items efficiently.
16. The computer-implemented method of claim 12 , where the one or more constraints further specifies a must-promote set of items, wherein if one item in the must-promote set is assigned a promotion price in a given period, all other items in the must-promote set must also be assigned a promotion price during a same period.
This invention relates to automated pricing and promotion systems for retail or e-commerce platforms, addressing the challenge of managing complex pricing rules while ensuring compliance with business constraints. The method involves dynamically assigning promotion prices to items based on predefined constraints, including a must-promote set requirement. This constraint ensures that if any item within a designated group is promoted during a specific period, all other items in that group must also receive a promotion price during the same timeframe. The system enforces this rule to maintain consistency in pricing strategies, such as bundling or category-wide promotions, while optimizing revenue or inventory turnover. The method may also incorporate additional constraints, such as budget limits, minimum/maximum price thresholds, or competitive pricing rules, to further refine the promotion assignment process. The automated approach reduces manual effort and ensures adherence to business policies, improving operational efficiency and customer satisfaction. The solution is particularly useful for large-scale retailers or online marketplaces where manual price adjustments are impractical.
17. The computer-implemented method of claim 12 , where the one or more constraints further specifies a cannot-promote set of items, wherein if one item in the cannot-promote set is assigned a promotion price in a given period, none of the other items in the cannot-promote set may be assigned a promotion price during a same period.
This invention relates to dynamic pricing systems for retail or e-commerce platforms, specifically addressing the challenge of managing promotional pricing conflicts between competing products. The method involves assigning promotion prices to items while enforcing constraints to prevent undesirable overlaps. One key constraint is a "cannot-promote" set, which ensures that if any item within a defined group is promoted in a given time period, no other item in that group can receive a promotion during the same period. This prevents cannibalization of sales between similar or competing products, maintaining profitability and strategic pricing control. The system dynamically evaluates pricing decisions based on predefined rules, historical data, and real-time demand, optimizing revenue while adhering to business constraints. The method may also incorporate additional constraints, such as budget limits, minimum/maximum price thresholds, or competitive pricing rules, to further refine promotion strategies. The solution is particularly useful for retailers and online marketplaces where automated pricing adjustments are critical for maximizing sales without undermining product margins or customer perception.
18. The computer-implemented method of claim 12 , where the one or more constraints further comprise a fixed relationship between prices of given items in the group that must be maintained in all periods.
This invention relates to dynamic pricing systems for groups of items, addressing the challenge of maintaining consistent pricing relationships between items over time while optimizing revenue. The method involves adjusting prices for a group of items across multiple periods, where the adjustments are constrained by predefined relationships between the prices of specific items. These constraints ensure that the price of one item in the group remains fixed relative to another, regardless of external factors or changes in demand. The system dynamically updates prices while enforcing these constraints, allowing for flexible pricing strategies that adapt to market conditions without disrupting established pricing hierarchies. The method also incorporates additional constraints, such as minimum or maximum price thresholds, to further control pricing behavior. By maintaining these fixed relationships, the system ensures fairness, consistency, and predictability in pricing, which is particularly useful in industries where relative pricing is critical, such as retail, subscription services, or bundled offerings. The approach balances revenue optimization with the need to preserve pricing structures that customers or stakeholders rely on.
Unknown
August 11, 2020
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.